|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2006.06.3560 on September 25 2006 © 2006 American Society of Clinical Oncology. Statins and Cancer Risk: A Literature-Based Meta-Analysis and Meta-Regression Analysis of 35 Randomized Controlled Trials
From the Departments of Pharmacology and Pathophysiology, School of Medicine, University of Athens; and the Department of Epidemiological Surveillance & Intervention, Hellenic Center for Disease Control & Prevention, Athens, Greece Address reprint requests to Stefanos Bonovas, MD, MSc, Department of Pharmacology, School of Medicine, University of Athens, 75 Mikras Asias Str, Athens 11527, Greece; e-mail: sbonovas{at}med.uoa.gr
PURPOSE: A growing body of literature suggests that statins may have chemopreventive potential against cancer. Our aim was to examine the strength of this association through a detailed meta-analysis and meta-regression analysis of randomized controlled trials (RCTs). METHODS: A comprehensive search for trials published up to 2005 was performed, reviews of each study were conducted, and data were abstracted. Before meta-analysis, the studies were evaluated for publication bias and heterogeneity. Pooled relative risk (RR) estimates and 95% CIs were calculated using the random- and fixed-effects models. Subgroup, sensitivity, and meta-regression analyses were also conducted. RESULTS: Thirty-five RCTs of statins for cardiovascular outcomes contributed to the analysis (n = 109,143). The degree of variability between trials was consistent with what would be expected to occur by chance alone. Statin use was not associated with a substantially increased or decreased overall risk of cancer (RR = 0.99; 95% CI, 0.94 to 1.04). Similarly, statin use did not significantly affect respiratory cancer risk (RR = 0.95; 95% CI, 0.83 to 1.09). However, the meta-regression analysis indicated that age of study participants modified the association between statin use and cancer risk (P = .003). CONCLUSION: Our findings do not support a protective effect of statins against cancer. However, this conclusion is limited by the relatively short follow-up periods (4.5 years on average) of the studies analyzed. Thus, it is important to continue monitoring the long-term safety profiles of statins. Until then, physicians need to be vigilant in ensuring that statin use remains restricted to the approved indications.
Statins comprise a therapeutic class of agents that reduce plasma cholesterol levels by inhibiting hepatic hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase, the rate-controlling enzyme in cholesterol synthesis.1 Originally approved to prevent cardiovascular disease, statins have now been suggested to have some efficacy in an increasing number of diseases, including dementia,2-4 multiple sclerosis,5-7 and rheumatoid arthritis,8,9 and fractures.10-13 A growing body of evidence also suggests that statins may have chemopreventive potential against cancer.14-16 Well-designed epidemiologic studies indicate considerable reductions of overall cancer risk among statin users.17-19 Furthermore, recent observational studies link statins with beneficial effects in site-specific cancers, including colorectal,20 breast,21 prostate,22 lung,23 and pancreatic cancers.24 However, none of these findings has been confirmed in large randomized controlled trials (RCTs), and these data cannot thus be taken as evidence that statins prevent cancer. On the other hand, several RCTs designed to assess the safety of statins and cardiovascular outcomes among patients receiving them have also assessed the incidence of cancer. However, because these studies were designed to assess end points related to cardiovascular disease, the small numbers of cancers observed limit their power to detect associations between statin use and cancer risk. In the past, two meta-analyses of RCTs found no evidence that statin use may affect cancer risk.25,26 However, these reports were based on a limited number of trials and a small number of patients, and did not include the large RCTs of statins published subsequently. Given the widespread and long-term use of statins, more knowledge is needed on the relationship between these medications and cancer. To address this issue, we conducted a detailed meta-analysis and meta-regression analysis of RCTs published in peer-reviewed literature.
Search Strategy To identify the studies of interest we searched the MEDLINE (1966 through July 2005) and SCI Expanded (1970 through July 2005) databases. Search terms included "HMG-CoA reductase inhibitor(s)," "statin(s)," "atorvastatin," "cerivastatin", "fluvastatin," "lovastatin," "mevastatin," "pravastatin," "rivastatin," "rosuvastatin," and "simvastatin." The search was limited to RCTs with human subjects. The title and abstract of studies identified in the computerized search were scanned to exclude any that were clearly irrelevant. The full text of the remaining articles was read to determine whether they contained information on the topic of interest. The reference lists of the articles were reviewed to identify citations to other studies of the same topic.
Selection Criteria We did not assess the methodologic quality of the primary studies, since quality assessment in meta-analysis is controversial. Scores constructed in an ad hoc fashion may lack demonstrated validity, and results may not be associated with quality.27,28 Instead, we performed subgroup and sensitivity analyses as it is recommended.29,30
Data Extraction Risk ratios and 95% CIs were estimated by reconstructing contingency tables based on the number of patients randomly assigned and the number of patients experiencing cancer events. When no outcomes occurred in one group, 1 was added to each cell of the respective contingency table. Nonmelanoma skin cancers were excluded from any analysis. Differences in data extraction were resolved by consensus, referring back to the original article.
Meta-Analysis
Publication bias was evaluated using the funnel graph, the Begg-Mazumdar adjusted rank correlation test,34 and the Egger et al35 regression asymmetry test. To evaluate whether the results of the studies were homogeneous, we used the Cochran's Q test.36 It is a Subgroup analyses were also performed according to the type of statin. This was done to investigate potentially different effects on risk. We then performed sensitivity analysis by restricting the meta-analysis: (a) to trials that evaluated a statin therapy compared with placebo, (b) to trials with a minimum duration of 3 years, (c) to trials that enrolled at least 3,000 subjects, and (d) to trials that fulfilled both (b) and (c) criteria. By restricting the analysis to major trials, we attempted to make our results free of publication bias.39 Last, we performed a site-specific subgroup analysis to evaluate the association between statin use and respiratory cancer, which accounts for more than 15% of all cancer cases and 30% of all cancer deaths. Randomized trials that enrolled at least 3,000 participants and had a minimum duration of 3 years were selected to minimize publication bias because publication of respiratory cancer data of such large studies is less likely to depend on the magnitude and direction of their results.39 To evaluate the stability of the results, we also performed the leave-one-out sensitivity analysis. The scope of this analysis was to evaluate the influence of individual studies, by estimating the average RR in the absence of each study.
Meta-Regression Analysis We first converted all risk ratios by logarithmic transformations to achieve more symmetric distributions. Linear regression analysis, weighted by the inverse of the variance of the logarithm of the risk ratio, was used for the analysis. Before the analysis, the distribution of the weight of the studies was investigated to identify and exclude studies with extreme outlier weights, because they would dominate the analysis with their own characteristics. First, univariate linear regression analyses were performed for each factor. All factors were then included in a multivariate regression analysis with a backward stepwise approach to selecting the significant ones. The estimated coefficients of the weighted linear regression correspond to differences in the log risk ratios for one unit of difference in the explanatory factor. Thus, corresponding ratios of the risk ratios were obtained by exponentiation of these estimated coefficients. All tests are two tailed. For all tests, a probability level less than .05 was considered significant. This work was performed according to the QUOROM recommendations for meta-analyses of RCTs.40 Stata 6 (STATA Corporation, College Station, TX) software was used for the statistical analyses.
Search Results One thousand six hundred fifty-two records were identified by MEDLINE search, and 407 by the SCI database. However, most abstracts did not specifically address the topic of our analysis and were excluded from full-text review. We retrieved 103 potentially relevant manuscripts. The full text was read and the reference lists were checked. Finally, we identified 35 RCTs of statins for cardiovascular outcomes that met our inclusion criteria and contained primary data regarding the association between use of statins and overall cancer risk.41-75 So, we were able to conduct a post hoc analysis of these trials and calculate risk ratios for cancer. Twenty nine of 35 RCTs were trials of monotherapy with a statin compared with placebo,41,42,44-49,51-55,58-60,62-67,69-75 whereas six RCTs compared statin treatment with a usual care control group.43,50,56,57,61,68 Atorvastatin had been evaluated in six trials,41,43,44,47,49,57 fluvastatin in four,48,53,58,60 lovastatin in five,59,70,71,74,75 pravastatin in 14,45,50,52,54,56,62-69,73 and simvastatin in six.42,46,51,55,61,72 Thirty trials41-43,45,46,48,50-60,62-67,69-75 reported the overall incidence of cancer in both statin and control group. Cancer mortality was the only relevant outcome measure provided for the remaining five trials.44,47,49,61,68 The publication dates of the trials included in the meta-analysis ranged between 1991 and 2005. Table 1 lists the trials used in the meta-analysis together with the respective trial drug, number and summary characteristics of patients, duration of follow-up and estimated risk ratios and 95% CIs.
Meta-Analysis of Statin Use and Overall Cancer Risk Thirty-five RCTs of statins contributed to the analysis.41-75 A total of 109,143 individuals (females, approximately 26%) participated in these trials, with an average follow-up of nearly 4.5 years. The number of participants in individual trials ranged from 72 to 20,536 and the follow-up times ranged from 0.5 to 10.4 years (Table 1). The participants had a mean age of 61 years at enrollment. The proportion of subjects who experienced a cancer event during the follow-up was 5.91% (6,447 cancers), corresponding to an incidence rate of 0.0132 per year. Twenty-three trials reported a lower risk of cancer in the treatment group, 10 trials reported a higher risk, and two trials reported no association (RR = 1.0; Table 1). Only one trial, reporting a risk ratio higher than 1.0, had CI that did not include unity.54 Meta-analysis of all 35 trials showed no evidence for an association between statin therapy and overall cancer risk. The calculated effect estimate was identical either assuming a fixed- or a random-effects model (RR = 0.99; 95% CI, 0.94 to 1.04; P = .66; Table 2). Figure 1 graphs the risk ratios and 95% CIs from the individual trials and the pooled results.
The Cochran's Q test had a P value of .73 (Q = 28.61 on 34 df) and the corresponding quantity I2 was 0%, both indicating that the degree of variability between trials was consistent with what would be expected to occur by chance alone (Table 2). The P value for the Begg-Mazumdar test was .57. In contrast, the Egger test had a marginal P value of .07, suggestive of a possible bias. Indeed, the funnel plot did not have the expected shape. The right corner of the pyramidal part of the funnel, which should contain smaller trials reporting relative risks greater than 1, was missing (Fig 2).
When the analysis was restricted to trials that evaluated a statin therapy compared with placebo (n = 29), the calculated effect estimate did not change at all (Table 2). After stratifying the data into subgroups, according to the type of statin, we did not find any statistically significant association between atorvastatin, fluvastatin, lovastatin, pravastatin, or simvastatin use and cancer. However, newer statins (atorvastatin, fluvastatin) appeared to have lower RR estimates, though failed to reach statistical significance (Table 2). Then, we performed a heterogeneity test to compare RR across the five statin subgroups. It had a P value of .23, indicating that the degree of variability between the five predefined groups was consistent with what would be expected to occur by chance alone. When the analysis was restricted to trials with a minimum duration of 3 years (22 trials), we found no evidence of an association between statin use and cancer risk (RR = 0.99; 95% CI, 0.95 to 1.04). Similarly, when only trials that enrolled at least 3,000 subjects contributed to the analysis (11 trials), the pooled effect estimate suggested that an assumption of no association between statin use and cancer risk is reasonable (RR = 1.01; 95% CI, 0.95 to 1.07; Table 2). Last, we performed a meta-analysis of RCTs that fulfilled both previous criteria. Nine major trials contributed to this analysis; Scandinavian Simvastatin Survival Study (4S),42 Anglo-Scandinavian Cardiac Outcomes Trial–Lipid Lowering Arm (ASCOT-LLA),47 Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT),50 MRC/BHF,51 Long-term Intervention with Pravastatin in Ischaemic Disease (LIPID),52 Prospective Study of Pravastatin in the Elderly at Risk (PROSPER),54 Air Force Coronary Atherosclerosis Prevention Study (AFCAPS),59 Cholesterol and Reccurent Events (CARE)62 and West of Scotland Coronary Prevention Study (WOSCOPS). 69 These trials had enrolled approximately 78,000 individuals, with an average follow-up of nearly 5.3 years. Once again, we found no association between statin use and cancer risk either assuming a fixed- (RR = 1.01; 95% CI, 0.96 to 1.06) or a random-effects model (RR = 1.01; 95% CI, 0.96 to 1.07). In this restricted analysis, there was no evidence of publication bias (Begg-Mazumdar P = .75; Egger P = .69) or heterogeneity (Cochran's Q test P = .34; I2 = 11%; Table 2).
Meta-Analysis of Statin Use and Respiratory Cancer Risk The number of participants in the individual trials ranged from 4,444 to 20,536, and the follow-up times ranged from 3.2 to 10.4 years. The participants had a mean age of 63 years at enrollment. The number of incident respiratory cancer cases ranged from 39 to 346. Table 3 lists the trials used in the meta-analysis together with the estimated risk ratios and 95% CIs.
A total of 63,353 individuals (females, approximately 25%) participated in these randomized trials: 31,669 in treatment groups and 31,684 in control groups. They had an average follow-up of nearly 5.5 years. Eight hundred seventy-four subjects (1.38%) experienced a respiratory cancer event during follow-up, corresponding to an incidence rate of 0.0025 per year. Four trials42,50,52,69 reported a lower risk of respiratory cancer in the treatment group, whereas the other three trials51,54,59 reported a higher risk (Table 3). None was statistically significant. The overall incidence of respiratory cancer was 1.35% in the treatment group (426 incident respiratory cancer cases) and 1.41% in the control group (448 incident cases). Meta-analysis of all seven trials showed no evidence for an association between statin therapy and respiratory cancer risk. The calculated effect estimate was identical either assuming a fixed- or assuming a random-effects model (RR = 0.95; 95% CI, 0.83 to 1.09; n = 7). Figure 3 graphs the risk ratios and 95% CIs from the individual trials and the pooled results. The Cochran's Q test had a P value of .42 (Q = 6.01 on 6 df), and the corresponding quantity I2 was 0%, both indicating little variability between studies that cannot be explained by chance.
The funnel plot had the expected funnel shape (Fig 4) showing no evidence of publication bias. Furthermore, the P values for the Begg-Mazumdar test and Egger test were .55 and .84 respectively, both suggesting that an assumption of no publication bias is reasonable.
To evaluate the stability of the results, we also performed the leave-one-out sensitivity analysis. In this analysis, the overall effect size was calculated, removing one study at a time. This analysis confirmed the stability of our results (Table 4).
Meta-Regression Analysis To investigate the impact of study characteristics on the study estimates of relative risk, we performed a meta-regression analysis. One study51 was initially excluded from the analysis because it had an extreme outlier weight of 436, compared with a mean weight of 49 of the 34 remaining studies. The risk ratio of cancer for statin therapy varied between 0.25 and 1.77 across the studies and was 1.01 in the excluded study. Table 5 shows the meta-regression results. In the univariate linear regression analysis, the mean age of participants was the only variable statistically significantly associated with the risk ratio of cancer. A 10-year rise in the mean age of enrolled individuals increased the risk ratio by 14% (95% CI, 5% to 23%). The multivariate backward stepwise analysis confirmed this finding. Besides the mean age of participants at enrollment (ratio of RR = 1.14; 95% CI, 1.05 to 1.23; P = .003), all other variables failed to reach statistical significance. Figure 5 displays risk ratio estimates according to the mean age of study participants. It shows statin therapy to be associated with increasing risk ratios of cancer as the age of study participants rises.
To investigate the stability of this finding, we performed a sensitivity analysis. First, we ran the analysis after inclusion of the study with the extreme outlier weight.51 The results did not change (ratio of RR = 1.14; 95% CI, 1.05 to 1.23). Second, we removed from the analysis five studies45,49,61,68,73 that, though they had the largest variance (wide intervals in Fig 1), could exert much influence, because their age-RR data points were extreme. Once again, the results did not change substantially (ratio of RR = 1.13; 95% CI, 1.04 to 1.23). Last, we reran the meta-regression analysis after deleting the PROSPER study.54 The results remained in the same direction but did not reach statistical significance (ratio of RR = 1.08; 95% CI, 0.97 to 1.21).
There is a long-standing debate concerning the association between use of statins and cancer. In a review of rodent carcinogenicity tests, Newman and Hulley reported that lipid-lowering drugs, including statins, initiate or promote cancer in rats and mice.76 In contrast, several laboratory and epidemiologic studies indicate that statins may have chemopreventive potential against cancer.14-19 The possible growth-suppressing properties of the statins are supposed to be a result of the inhibition of HMG-CoA reductase, but other mechanisms have also been suggested.77 Furthermore, recent observational studies have found large reductions in the risk of site-specific cancers20-24 with the use of statin therapy. Meta-analysis of randomized trials allows for a more objective appraisal of the evidence, which may lead to resolution of uncertainty and disagreement. It serves as a valuable tool for studying rare and unintended effects of a treatment, by permitting synthesis of data and providing more stable estimates of effect. Our meta-analysis did not provide evidence that statin use is associated with a decreased or increased risk of cancer (RR = 0.99; 95% CI, 0.94 to 1.04; n = 35). When the analysis was restricted to major RCTs, with an average follow-up of nearly 5.3 years, the results did not change substantially (RR = 1.01; 95% CI, 0.96 to 1.06; n = 9). Similarly, it did not find evidence that statin use is associated with a substantially increased or decreased risk of respiratory cancer (RR = 0.95; 95% CI, 0.83 to 1.09; n = 7). Our findings are also supported by a recent site-specific meta-analysis on the association between use of statins and breast cancer.78 Likewise, it concluded that statin use does not substantially affect breast cancer risk. There is a clear disparity between the results provided by the previously mentioned observational studies and the results of our meta-analysis of 35 RCTs. Although randomized and observational designs sometimes produce equivalent results,79,80 systematic reviews have found that randomized and observational studies often give different results, and that the difference is in all directions.81 A likely explanation is that associations between statin use and cancer in observational studies are confounded by social and behavioral factors. Even when statistical adjustments are made for potential confounders, this inadequately captures the full extend of the complex ways in which social and behavioral factors confound associations between statins and cancer. Another explanation might be that randomized trials' entry criteria may account for disparities with epidemiologic studies. We also draw attention to the results of the meta-regression analysis. It indicated age of study participants to be associated with the RR of cancer. A 10-year rise in the mean age of enrolled individuals increased the risk ratio of cancer by 14% (P = .003). It implies that statin use is associated with a decreased risk of cancer in younger patients and an increased risk of cancer among older patients. This finding is in apparent agreement with the findings of a subgroup analysis of the LIPID trial.52 This analysis82 revealed that among older patients (age 65 to 75 years; n = 3,514) those assigned to pravastatin developed cancer more often (RR = 1.14; 95% CI, 0.98 to 1.32), whereas among younger patients (age 31 to 64 years; n = 5,500) those assigned to placebo developed cancer more often (RR = 0.87; 95% CI, 0.73 to 1.02; test for heterogeneity, P = .012). Furthermore, our finding justifies the results of the PROSPER trial,54 which is the only RCT that examined the efficacy and safety of a statin in a unique population of elderly individuals (age 70 to 82 years; n = 5,804). It detected a statistically significant imbalance in new cancer diagnoses, which was 25% more frequent in the pravastatin group (hazard ratio = 1.25; 95% CI, 1.04 to 1.51; P = .020). Our finding brings up again concerns about the safety of statins among elderly patients. The potential for increased risk of cancer with the lowering of cholesterol has initially arisen from the finding of an inverse association between plasma cholesterol and cancer rates, especially in older patients,83 and from the results of other studies.84-86 However, meta-regression describes observational relationships across studies, which are subject to confounding by other characteristics that vary between the trials. Even though every trial is randomized, meta-regression is only the study of the epidemiology of trials, and relations may well not be causal.87 Therefore, further verification is needed, and this finding should be interpreted with caution. When meta-analysis of published literature is performed, consideration of study bias is critical. Existence of a bias in favor of publication of statistically significant results is well documented in the literature,88,89 and it can lead to inflated estimates of effect in meta-analyses. In our study, the meta-analysis of all 35 trials provided an RR of 0.99, but the funnel plot did not have the expected shape and the Egger test (P = .07) was suggestive of publication bias. Indeed, the right corner of the pyramidal part of the funnel, which should contain smaller trials reporting relative risks greater than 1, was missing. However, when the analysis was restricted to the nine major trials (technique to minimize publication bias), the RR increased slightly to 1.01 and the Egger test was clearly not significant. So, we believe that the phenomenon of publication bias has been properly handled, and accounts for the little difference of RR (from 0.99 to 1.01). In contrast, the tests of heterogeneity indicated little variability between studies that cannot be explained by chance. Nevertheless, our meta-analysis has several limitations. First, our search was restricted to published studies. We did not search for unpublished trials or original data. Second, it included trials of statins in varying populations regarding age and subsequent cancer risk. Third, despite the differences between studies with respect to drugs and doses administered, all drugs (atorvastatin, fluvastatin, lovastatin, pravastatin, and simvastatin) have been regarded as being the same. Pharmacologically, this is not correct and may, therefore, have different effects on risk. Fourth, a main issue remaining beyond our control is cancer latency. Because the exposure and follow-up times lasted only for nearly 5 years, estimates of cancer risk resulting from longer exposure to statins are not possible. In conclusion, overview of existing data does not support a potential role of statins in cancer chemoprevention. Given the substantial proportion of the population using statins, the indications for long-term and perhaps lifelong use, it is important to continue monitoring their long-term safety profiles. Until then, physicians need to be vigilant in ensuring that use of statins remains restricted to the approved indications. Under no circumstances should statin use undermine the importance of healthy lifestyle habits, the merits of which are well established in cancer prevention.
The authors indicated no potential conflicts of interest.
published online ahead of print at www.jco.org on September 25, 2006. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Sewester CS, Dombek CE, Olin BR, et al: Drug facts and Comparisons, 2001 ed. St Louis, MO: Facts and Comparisons, 2001 2. Zamrini E, McGwin G, Roseman JM: Association between statin use and Alzheimer's disease. Neuroepidemiology 23:94-98, 2004[CrossRef][Medline] 3. Jick H, Zornberg GL, Jick SS, et al: Statins and the risk of dementia. Lancet 356:1627-1631, 2000[CrossRef][Medline] 4. Wolozin B, Kellman W, Ruosseau P, et al: Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors. Arch Neurol 57:1439-1443, 2000 5. Vollmer T, Key L, Durkalski V, et al: Oral simvastatin treatment in relapsing-remitting multiple sclerosis. Lancet 363:1607-1608, 2004[CrossRef][Medline] 6. Sena A, Pedrosa R, Graca Morais M: Therapeutic potential of lovastatin in multiple sclerosis. J Neurol 250:754-755, 2003[CrossRef][Medline] 7. Neuhaus O, Strasser-Fuchs S, Fazekas F, et al: Statins as immunomodulators: Comparison with interferon-beta 1b in MS. Neurology 59:990-997, 2002 8. McCarey DW, McInnes IB, Madhok R, et al: Trial of Atorvastatin in Rheumatoid Arthritis (TARA): Double-blind, randomised placebo-controlled trial. Lancet 363:2015-2021, 2004[CrossRef][Medline] 9. Kanda H, Hamasaki K, Kubo K, et al: Antiinflammatory effect of simvastatin in patients with rheumatoid arthritis. J Rheumatol 29:2024-2026, 2002[Medline] 10. Rejnmark L, Olsen ML, Johnsen SP, et al: Hip fracture risk in statin users—A population-based Danish case-control study. Osteoporos Int 15:452-458, 2004[Medline] 11. Chan KA, Andrade SE, Boles M, et al: Inhibitors of hydroxymethylglutaryl-coenzyme A reductase and risk of fracture among older women. Lancet 355:2185-2188, 2000[CrossRef][Medline] 12. Wang PS, Solomon DH, Mogun H, et al: HMG-CoA reductase inhibitors and the risk of hip fractures in elderly patients. JAMA 283:3211-3216, 2000 13. Meier CR, Schlienger RG, Kraenzlin ME, et al: HMG-CoA reductase inhibitors and the risk of fractures. JAMA 283:3205-3210, 2000 14. Graaf MR, Richel DJ, van Noorden CJ, et al: Effects of statins and farnesyltransferase inhibitors on the development and progression of cancer. Cancer Treat Rev 30:609-641, 2004[CrossRef][Medline] 15. Chan KK, Oza AM, Siu LL: The statins as anticancer agents. Clin Cancer Res 9:10-19, 2003 16. Jakobisiak M, Golab J: Potential antitumor effects of statins. Int J Oncol 23:1055-1069, 2003[Medline] 17. Friis S, Poulsen AH, Johnsen SP, et al: Cancer risk among statin users: A population-based cohort study. Int J Cancer 114:643-647, 2005[CrossRef][Medline] 18. Graaf MR, Beiderbeck AB, Egberts AC, et al: The risk of cancer in users of statins. J Clin Oncol 22:2388-2394, 2004 19. Blais L, Desgagne A, LeLorier J: 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors and the risk of cancer: A nested case-control study. Arch Intern Med 160:2363-2368, 2000 20. Poynter JN, Gruber SB, Higgins PD, et al: Statins and the risk of colorectal cancer. N Engl J Med 352:2184-2192, 2005 21. Cauley JA, Zmuda JM, Lui LY, et al: Lipid-lowering drug use and breast cancer in older women: A prospective study. J Womens Health (Larchmt) 12:749-756, 2003[Medline] 22. Shannon J, Tewoderos S, Garzotto M, et al: Statins and prostate cancer risk: A case-control study. Am J Epidemiol 162:318-325, 2005 23. Khurana V, Kochhar R, Bejjanki HR, et al: Statins reduce the incidence of lung cancer: A study of half a million US Veterans. J Clin Oncol 23:107s, 2005 (suppl; abstr 1006) 24. Khurana V, Chico G, Jaganmohan S, et al: Statins reduce the risk of pancreatic cancer in humans: Half a million veterans' case-control study. Pancreas 29:368, 2004 (abstr) 25. Hebert PR, Gaziano JM, Chan KS, et al: Cholesterol lowering with statin drugs, risk of stroke, and total mortality: An overview of randomized trials. JAMA 278:313-321, 1997[Abstract] 26. Bjerre LM, LeLorier J: Do statins cause cancer? A meta-analysis of large randomized clinical trials. Am J Med 110:716-723, 2001[CrossRef][Medline] 27. Greenland S: Invited commentary. Am J Epidemiol 140:290-296, 1994 28. Juni P, Witschi A, Bloch R, et al: The hazards of scoring the quality of clinical trials for meta-analysis. JAMA 282:1054-1060, 1999 29. Hasselblad V, Eddy DM, Kotchmar DJ: Synthesis of environmental evidence. J Air Waste Manag Assoc 42:662-671, 1992 30. Friedenreich CM, Brant RF, Riboli E: Influence of methodologic factors in a pooled analysis of 13 case-control studies of colorectal cancer and dietary fiber. Epidemiology 5:66-67, 1994[Medline] 31. Mantel N, Haenszel W: Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22:719-748, 1959[Medline] 32. DerSimonian R, Laird N: Meta-analysis in clinical trials. Control Clin Trials 7:177-188, 1986[CrossRef][Medline] 33. Petitti DB: Statistical methods in meta-analysis, in Petitti DB (ed): Meta-analysis, Decision Analysis and Cost-Effectiveness Analysis. New York, NY, Oxford University Press, 1994, pp 90-114 34. Begg CB, Mazumdar M: Operating characteristics of a rank correlation test for publication bias. Biometrics 50:1088-1101, 1994[CrossRef][Medline] 35. Egger M, Smith GD, Schneider M, et al: Bias in meta-analysis detected by a simple graphical test. BMJ 315:629-634, 1997 36. Cochran WG: The combination of estimates from different experiments. Biometrics 8:101-129, 1954[Medline] 37. Higgins JPT, Thompson SG: Quantifying heterogeneity in a meta-analysis. Stat Med 21:1539-1558, 2002[CrossRef][Medline] 38. Higgins JPT, Thompson SG, Deeks JJ, et al: Measuring inconsistency in meta-analyses. BMJ 327:557-560, 2003 39. Felson DT: Bias in meta-analytic research. J Clin Epidemiol 45:885-892, 1992[CrossRef][Medline] 40. Moher D, Cook DJ, Eastwood S, et al: Improving the quality of reports of meta-analyses of randomised controlled trials: The QUOROM statement—Quality of Reporting of Meta-analyses. Lancet 354:1896-1900, 1999[CrossRef][Medline] 41. Wanner C, Krane V, Marz W, et al: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 353:238-248, 2005 42. Strandberg TE, Pyorala K, Cook TJ, et al: Mortality and incidence of cancer during 10-year follow-up of the Scandinavian Simvastatin Survival Study (4S). Lancet 364:771-777, 2004[CrossRef][Medline] 43. Koren MJ, Hunninghake DB: Clinical outcomes in managed-care patients with coronary heart disease treated aggressively in lipid-lowering disease management clinics: The Alliance study. J Am Coll Cardiol 44:1772-1779, 2004 44. Colhoun HM, Betteridge DJ, Durrington PN, et al: Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): Multicentre randomised placebo-controlled trial. Lancet 364:685-696, 2004[CrossRef][Medline] 45. Zanchetti A, Crepaldi G, Bond MG, et al: Different effects of antihypertensive regimens based on fosinopril or hydrochlorothiazide with or without lipid lowering by pravastatin on progression of asymptomatic carotid atherosclerosis: Principal results of PHYLLIS—a randomized double-blind trial. Stroke 35:2807-2812, 2004 46. Beishuizen ED, van de Ree MA, Jukema JW, et al: Two-year statin therapy does not alter the progression of intima-media thickness in patients with type 2 diabetes without manifest cardiovascular disease. Diabetes Care 27:2887-2892, 2004 47. Sever PS, Dahlof B, Poulter NR, et al: Prevention of coronary and stroke events with atorvastatin in hypertensive patients who have average or lower-than-average cholesterol concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–Lipid Lowering Arm (ASCOT-LLA): A multicentre randomised controlled trial. Lancet 361:1149-1158, 2003[CrossRef][Medline] 48. Holdaas H, Fellstrom B, Jardine AG, et al: Assessment of LEscol in Renal Transplantation (ALERT) Study Investigators: Effect of fluvastatin on cardiac outcomes in renal transplant recipients—A multicentre, randomised, placebo-controlled trial. Lancet 361:2024-2031, 2003[CrossRef][Medline] 49. Mohler ER III, Hiatt WR, Creager MA: Cholesterol reduction with atorvastatin improves walking distance in patients with peripheral arterial disease. Circulation 108:1481-1486, 2003 50. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial—Major outcomes in moderately hypercholesterolemic, hypertensive patients randomized to pravastatin vs usual care: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT). JAMA 288:2998-3007, 2002 51. Heart Protection Study Collaborative Group: MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial. Lancet 360:7-22, 2002[CrossRef][Medline] 52. LIPID Study Group (Long-term Intervention with Pravastatin in Ischaemic Disease): Long-term effectiveness and safety of pravastatin in 9014 patients with coronary heart disease and average cholesterol concentrations: The LIPID trial follow-up. Lancet 359:1379-1387, 2002[CrossRef][Medline] 53. Serruys PW, de Feyter P, Macaya C, et al: Fluvastatin for prevention of cardiac events following successful first percutaneous coronary intervention: A randomized controlled trial. JAMA 287:3215-3222, 2002 54. Shepherd J, Blauw GJ, Murphy MB, et al: Prospective Study of Pravastatin in the Elderly at Risk: Pravastatin in elderly individuals at risk of vascular disease (PROSPER)—A randomised controlled trial. Lancet 360:1623-1630, 2002[CrossRef][Medline] 55. Teo KK, Burton JR, Buller CE, et al: Long-term effects of cholesterol lowering and angiotensin-converting enzyme inhibition on coronary atherosclerosis: The Simvastatin/Enalapril Coronary Atherosclerosis Trial (SCAT). Circulation 102:1748-1754, 2000 56. GISSI Prevenzione Investigators: Results of the low-dose (20 mg) pravastatin GISSI Prevenzione trial in 4271 patients with recent myocardial infarction: Do stopped trials contribute to overall knowledge? Ital Heart J 1:810-820, 2000[Medline] 57. Pitt B, Waters D, Brown WV, et al: Aggressive lipid-lowering therapy compared with angioplasty in stable coronary artery disease: Atorvastatin versus Revascularization Treatment Investigators. N Engl J Med 341:70-76, 1999 58. Serruys PW, Foley DP, Jackson G, et al: A randomized placebo-controlled trial of fluvastatin for prevention of restenosis after successful coronary balloon angioplasty: Final results of the Fluvastatin Angiographic Restenosis (FLARE) trial. Eur Heart J 20:58-69, 1999 59. Downs JR, Clearfield M, Weis S, et al: Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: Results of AFCAPS/TexCAPS—Air Force/Texas Coronary Atherosclerosis Prevention Study. JAMA 279:1615-1622, 1998 60. Herd JA, Ballantyne CM, Farmer JA, et al: Effects of fluvastatin on coronary atherosclerosis in patients with mild to moderate cholesterol elevations (Lipoprotein and Coronary Atherosclerosis Study [LCAS]). Am J Cardiol 80:278-286, 1997[CrossRef][Medline] 61. Wenke K, Meiser B, Thiery J, et al: Simvastatin reduces graft vessel disease and mortality after heart transplantation: A four-year randomized trial. Circulation 96:1398-1402, 1997 62. Sacks FM, Pfeffer MA, Moye LA, et al: The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels: Cholesterol and Recurrent Events Trial investigators. N Engl J Med 335:1001-1009, 1996 63. Mercuri M, Bond MG, Sirtori CR, et al: Pravastatin reduces carotid intima-media thickness progression in an asymptomatic hypercholesterolemic mediterranean population: The Carotid Atherosclerosis Italian Ultrasound Study. Am J Med 101:627-634, 1996[CrossRef][Medline] 64. Jukema JW, Bruschke AV, van Boven AJ, et al: Effects of lipid lowering by pravastatin on progression and regression of coronary artery disease in symptomatic men with normal to moderately elevated serum cholesterol levels: The Regression Growth Evaluation Statin Study (REGRESS). Circulation 91:2528-2540, 1995 65. Salonen R, Nyyssonen K, Porkkala E, et al: A population-based primary preventive trial of the effect of LDL lowering on atherosclerotic progression in carotid and femoral arteries. Circulation 92:1758-1764, 1995 66. Pitt B, Mancini GB, Ellis SG, et al: Pravastatin limitation of atherosclerosis in the coronary arteries (PLAC-I): Reduction in atherosclerosis progression and clinical events—PLAC-I investigation. J Am Coll Cardiol 26:1133-1139, 1995[Abstract] 67. Crouse JR 3rd, Byington RP, Bond MG, et al: Pravastatin, Lipids, and Atherosclerosis in the Carotid Arteries (PLAC-II). Am J Cardiol 75:455-459, 1995[CrossRef][Medline] 68. Kobashigawa JA, Katznelson S, Laks H, et al: Effect of pravastatin on outcomes after cardiac transplantation. N Engl J Med 333:621-627, 1995 69. Shepherd J, Cobbe SM, Ford I, et al: Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia: West of Scotland Coronary Prevention Study Group. N Engl J Med 333:1301-1307, 1995 70. Furberg CD, Adams HP Jr, Applegate WB, et al: Effect of lovastatin on early carotid atherosclerosis and cardiovascular events: Asymptomatic Carotid Artery Progression Study (ACAPS) Research Group. Circulation 90:1679-1687, 1994 71. Waters D, Higginson L, Gladstone P, et al: Effects of monotherapy with an HMG-CoA reductase inhibitor on the progression of coronary atherosclerosis as assessed by serial quantitative arteriography: The Canadian Coronary Atherosclerosis Intervention Trial. Circulation 89:959-968, 1994 72. MAAS Investigators: Effect of simvastatin on coronary atheroma: The Multicentre Anti-Atheroma Study (MAAS). Lancet 344:633-638, 1994[CrossRef][Medline] 73. The Pravastatin Multinational Study Group for Cardiac Risk Patients: Effects of pravastatin in patients with serum total cholesterol levels from 5.2 to 7.8 mmol/liter (200 to 300 mg/dl) plus two additional atherosclerotic risk factors. Am J Cardiol 72:1031-1037, 1993[CrossRef][Medline] 74. Blankenhorn DH, Azen SP, Kramsch DM, et al: Coronary angiographic changes with lovastatin therapy: The Monitored Atherosclerosis Regression Study (MARS). Ann Intern Med 119:969-976, 1993 75. Bradford RH, Shear CL, Chremos AN, et al: Expanded Clinical Evaluation of Lovastatin (EXCEL) study results: I, Efficacy in modifying plasma lipoproteins and adverse event profile in 8245 patients with moderate hypercholesterolemia. Arch Intern Med 151:43-49, 1991[Abstract] 76. Newman TB, Hulley SB: Carcinogenicity of lipid-lowering drugs. JAMA 275:55-60, 1996[Abstract] 77. Rao S, Porter DC, Chen X, et al: Lovastatin-mediated G1 arrest is through inhibition of the proteasome, independent of hydroxymethyl glutaryl-CoA reductase. Proc Natl Acad Sci U S A 96:7797-7802, 1999 78. Bonovas S, Filioussi K, Tsavaris N, et al: Use of statins and breast cancer: A meta-analysis of seven randomized clinical trials and nine observational studies. J Clin Oncol 23:8606-8612, 2005 79. Concato J, Shah N, Horwitz RI: Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 342:1887-1892, 2000 80. Benson K, Hartz AJ: A comparison of observational studies and randomized, controlled trials. N Engl J Med 342:1878-1886, 2000 81. Kunz R, Oxman AD: The unpredictability paradox: Review of empirical comparisons of randomised and non-randomised clinical trials. BMJ 317:1185-1190, 1998 82. Hunt D, Young P, Simes J, et al: Benefits of pravastatin on cardiovascular events and mortality in older patients with coronary heart disease are equal to or exceed those seen in younger patients: Results from the LIPID trial. Ann Intern Med 134:931-940, 2001 83. Weverling-Rijnsburger AW, Blauw GJ, Lagaay AM, et al: Total cholesterol and risk of mortality in the oldest old. Lancet 350:1119-1123, 1997[CrossRef][Medline] 84. Stemmermann GN, Chyou PH, Kagan A, et al: Serum cholesterol and mortality among Japanese American men: The Honolulu (Hawaii) Heart Program. Arch Intern Med 151:969-972, 1991[Abstract] 85. Schuit AJ, Van Dijk CE, Dekker JM, et al: Inverse association between serum total cholesterol and cancer mortality in Dutch civil servants. Am J Epidemiol 137:966-976, 1993 86. Schatzkin A, Hoover RN, Taylor PR, et al: Site-specific analysis of total serum cholesterol and incident cancer in the National Health and Nutrition Examination Survey I Epidemiologic Follow-Up Study. Cancer Res 48:452-458, 1988 87. Thompson SG, Higgins JP: How should meta-regression analyses be undertaken and interpreted? Stat Med 21:1559-1573, 2002[CrossRef][Medline] 88. Easterbrook PJ, Berlin JA, Gopalan R, et al: Publication bias in research. Lancet 337:867-872, 1991[CrossRef][Medline] 89. Simes JR: Publication bias: The case for an international registry of trials. J Clin Oncol 4:1529-1541, 1986 Submitted February 25, 2006; accepted June 16, 2006. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 test with df equal to the number of studies minus one, and tests the null hypothesis that the difference between the study estimates of RR is due to chance (the smaller the P value, the less homogeneity present among study results). We also calculated the quantity I2 
